WEBVTT 00:00:00.000 --> 00:00:02.060 Ever wish you had a super smart AI assistant 00:00:02.060 --> 00:00:05.000 living right inside your computer's terminal, 00:00:05.179 --> 00:00:07.679 completely free? Google just dropped something 00:00:07.679 --> 00:00:10.480 that could fundamentally change how pretty much 00:00:10.480 --> 00:00:12.259 everyone who writes code approaches their day. 00:00:12.380 --> 00:00:15.779 It's called Gemini CLI. Welcome back to the Deep 00:00:15.779 --> 00:00:17.699 Dive, everyone. We're here to unpack complex 00:00:17.699 --> 00:00:20.140 ideas into, you know, understandable nuggets. 00:00:20.460 --> 00:00:22.920 And today, yeah, we're diving into a tool that's 00:00:22.920 --> 00:00:25.420 been quietly reshaping the coding landscape. 00:00:25.620 --> 00:00:28.530 It's truly a game changer, I think, but... Maybe 00:00:28.530 --> 00:00:30.589 in ways you might not expect at first glance. 00:00:30.690 --> 00:00:32.710 We're talking about Google's Gemini CLI. Right. 00:00:32.810 --> 00:00:36.030 A free, powerful AI assistant that promises to, 00:00:36.170 --> 00:00:38.210 well, supercharge your development workflow. 00:00:38.469 --> 00:00:41.119 Exactly. For this deep dive, we've pulled from 00:00:41.119 --> 00:00:43.659 a really fascinating source, how to build an 00:00:43.659 --> 00:00:46.640 AI system that creates automated workflows. And 00:00:46.640 --> 00:00:48.960 we're specifically focusing on the core insights 00:00:48.960 --> 00:00:52.539 about Gemini CLI, our mission, to figure out 00:00:52.539 --> 00:00:54.920 if this tool truly deserves a permanent spot 00:00:54.920 --> 00:00:57.100 in your developable toolkit. And we'll explore 00:00:57.100 --> 00:01:00.479 the economics of its free tier, which are kind 00:01:00.479 --> 00:01:02.619 of mind -blowing, plus its standout features 00:01:02.619 --> 00:01:06.000 beyond just being a chatbot. And then, crucially... 00:01:06.359 --> 00:01:08.579 It's real world performance. We put it through 00:01:08.579 --> 00:01:10.739 the gauntlet, didn't we? We really did. And the 00:01:10.739 --> 00:01:13.719 results were, well, a mix. Some mind -blowing 00:01:13.719 --> 00:01:16.560 moments and some, frankly, deeply disappointing 00:01:16.560 --> 00:01:18.659 ones. We'll also give you the roadmap for getting 00:01:18.659 --> 00:01:20.920 started, reveal some advanced power features, 00:01:21.239 --> 00:01:23.980 size it up against competitors like GitHub Copilot, 00:01:24.079 --> 00:01:26.939 and look at the future of AI power development. 00:01:27.319 --> 00:01:29.439 So stick with us because this is where it gets 00:01:29.439 --> 00:01:31.840 really interesting. Okay, let's unpack this elephant 00:01:31.840 --> 00:01:34.920 in the room first. The pricing. For developers 00:01:34.920 --> 00:01:37.200 using personal Google accounts, the free tier 00:01:37.200 --> 00:01:40.359 for Gemini CLI is, well, it's incredibly generous. 00:01:40.659 --> 00:01:43.000 It really is. And you get access to Google's 00:01:43.000 --> 00:01:47.060 top tier, most advanced model, Gemini 2 .5 Pro. 00:01:47.280 --> 00:01:49.379 This isn't some dumbed down version. No, it's 00:01:49.379 --> 00:01:52.140 the real deal. And the limits. 60 requests per 00:01:52.140 --> 00:01:53.799 minute. That's one every second. Yeah. And a 00:01:53.799 --> 00:01:56.540 high daily limit to 2 ,000 requests per day. 00:01:56.599 --> 00:02:01.159 Plus that colossal context window. 1 million 00:02:01.159 --> 00:02:03.819 tokens. Which is hard to even visualize, right? 00:02:03.879 --> 00:02:05.959 It's like feeding it a very, very long book full 00:02:05.959 --> 00:02:08.319 of code. Exactly. It can hold a huge amount of 00:02:08.319 --> 00:02:10.360 project information in its head, so to speak. 00:02:10.599 --> 00:02:14.419 Now, for the mind -blowing math of free, a single 00:02:14.419 --> 00:02:17.340 request using that entire 1 million token context, 00:02:17.719 --> 00:02:21.460 normally that would cost around $3 on a standard 00:02:21.460 --> 00:02:24.120 pay -as -you -go AI plan. Right. Meaning you 00:02:24.120 --> 00:02:27.319 could theoretically use, what, $180 worth of 00:02:27.319 --> 00:02:30.050 AI processing power every single minute? Completely 00:02:30.050 --> 00:02:32.330 for free. It sounds almost too good to be true. 00:02:32.550 --> 00:02:34.490 Well, the author of our source material tested 00:02:34.490 --> 00:02:37.750 this heavily for days. Complex analysis, generating 00:02:37.750 --> 00:02:40.069 multiple files, really pushing the limits, and 00:02:40.069 --> 00:02:42.930 the total cost. Go on. Three cents. Seriously, 00:02:43.129 --> 00:02:45.990 that's not a typo. Just three cents. Wow. Okay. 00:02:46.210 --> 00:02:49.069 So what does this all mean? Google clearly wants 00:02:49.069 --> 00:02:51.610 Gemini to be the foundational tool for, like, 00:02:51.650 --> 00:02:53.629 the next generation of software development. 00:02:53.849 --> 00:02:55.250 It seems that way. They're willing to foot the 00:02:55.250 --> 00:02:57.819 bill to get it deep into your workflow. So probing 00:02:57.819 --> 00:03:00.419 question time, why is Google offering such a 00:03:00.419 --> 00:03:03.919 powerful tool essentially for free? They want 00:03:03.919 --> 00:03:06.060 Gemini to be the indispensable foundation for 00:03:06.060 --> 00:03:07.680 future development. And this isn't just another 00:03:07.680 --> 00:03:10.780 conversational AI. You know, Gemini CLI was built 00:03:10.780 --> 00:03:13.659 from the ground up as a developer native tool. 00:03:13.740 --> 00:03:16.199 It has features that really set it apart. Yeah, 00:03:16.240 --> 00:03:18.360 like the native Google search integration. You 00:03:18.360 --> 00:03:20.319 can ask it to search the web directly from your 00:03:20.319 --> 00:03:22.319 command line. No more flipping back and forth 00:03:22.319 --> 00:03:24.500 between the terminal and your browser. Right. 00:03:24.699 --> 00:03:27.639 Like, what is the latest stable version of React? 00:03:27.939 --> 00:03:30.680 Or find me the official docs for the express 00:03:30.680 --> 00:03:33.400 .js routing module. And the AI just gives you 00:03:33.400 --> 00:03:35.979 a clean, concise summary right there in the terminal. 00:03:36.360 --> 00:03:38.860 Keeps you in the zone. Totally. Then there's 00:03:38.860 --> 00:03:40.840 MCP integration. That's the Model Context Protocol, 00:03:41.120 --> 00:03:44.759 MCP. It's basically a way for the AI to securely 00:03:44.759 --> 00:03:47.919 connect to external services and tools you authorize. 00:03:48.039 --> 00:03:50.259 Think of MCP as like a secure bridge, right? 00:03:50.340 --> 00:03:52.639 Letting the AI interact with your digital world, 00:03:52.699 --> 00:03:55.240 but only the parts you allow. Exactly. Examples 00:03:55.240 --> 00:03:58.699 could be Context 7 for documentation, Notion 00:03:58.699 --> 00:04:02.199 for databases, Gmail, even GitHub for code analysis. 00:04:02.500 --> 00:04:04.560 And the setup sounds simple, like just an NPM 00:04:04.560 --> 00:04:07.439 install command. Pretty much. npm install dash 00:04:07.439 --> 00:04:10.240 g at context 7 missy p server is the example 00:04:10.240 --> 00:04:12.120 given. And then you just tell Gemini to use it 00:04:12.120 --> 00:04:13.919 in natural language. Very neat. And finally, 00:04:13.919 --> 00:04:16.740 maybe the most useful daily feature, intelligent 00:04:16.740 --> 00:04:20.339 file analysis. Oh yeah, this is big. Using the 00:04:20.339 --> 00:04:23.220 at symbol, you tell the AI to reference any file 00:04:23.220 --> 00:04:26.000 or folder in your project. So you could say at 00:04:26.000 --> 00:04:29.779 read e .md, summarize the main purpose, or even 00:04:29.779 --> 00:04:32.319 at circ components. Analyze all React components 00:04:32.319 --> 00:04:34.959 to just improvements. And it doesn't just read 00:04:34.959 --> 00:04:36.720 the text, does it? It understands the context 00:04:36.720 --> 00:04:39.439 of your entire code base, the dependencies, the 00:04:39.439 --> 00:04:41.699 relationships between files. It's like stacking 00:04:41.699 --> 00:04:44.699 Lego blocks of data, as the source put it, building 00:04:44.699 --> 00:04:46.959 a whole model of the project. So how does its 00:04:46.959 --> 00:04:49.779 file analysis specifically enhance coding casks 00:04:49.779 --> 00:04:52.439 day to day? It understands your entire code base, 00:04:52.519 --> 00:04:54.800 suggesting improvements based on the full project 00:04:54.800 --> 00:04:58.300 context. Okay, so theory is one thing. To really 00:04:58.300 --> 00:05:01.000 see what Gemini CLI could do, the author put 00:05:01.000 --> 00:05:03.519 it through three increasingly difficult real 00:05:03.519 --> 00:05:06.980 -world scenarios. Test one, the simple bug fix. 00:05:07.199 --> 00:05:09.620 And this was a clear pass, big green checkmark. 00:05:09.740 --> 00:05:11.660 The problem was a scrolling issue in a React 00:05:11.660 --> 00:05:14.300 app, content getting cut off. Described it in 00:05:14.300 --> 00:05:16.639 plain English. And the AI analyzed the code, 00:05:16.779 --> 00:05:20.040 found the CSS issue overflow, hidden was the 00:05:20.040 --> 00:05:22.829 culprit, and proposed a clean fix. Crucially, 00:05:22.889 --> 00:05:25.329 it showed the fix as a visual diff. You know, 00:05:25.389 --> 00:05:28.149 red for old code, green for new. Made it super 00:05:28.149 --> 00:05:30.110 easy to review. Yeah, and the whole thing took 00:05:30.110 --> 00:05:32.550 less than two minutes. Accurate diagnosis, professional 00:05:32.550 --> 00:05:36.209 fix, really seamless. Whoa! Imagine scaling that 00:05:36.209 --> 00:05:38.930 speed across thousands of components or tiny 00:05:38.930 --> 00:05:41.889 bugs. That's a true productivity multiplier right 00:05:41.889 --> 00:05:44.329 there. Definitely. Then test two, effortless 00:05:44.329 --> 00:05:47.990 unit test generation. This was another win. Seamless 00:05:47.990 --> 00:05:51.209 and impressive. The challenge. Write Jest unit 00:05:51.209 --> 00:05:54.089 tests for an existing utility file, utils .js. 00:05:54.509 --> 00:05:56.829 Jest helps you automatically check if your code 00:05:56.829 --> 00:05:59.250 works, basically. Simple prompt. Write basic 00:05:59.250 --> 00:06:01.850 Jest unit tests for the functions in utils .js. 00:06:02.189 --> 00:06:04.329 And within seconds, Gemini scanned it, identified 00:06:04.329 --> 00:06:07.089 the functions, and boom, proposed a new utils 00:06:07.089 --> 00:06:09.670 .test .js file. And it even explained what it 00:06:09.670 --> 00:06:12.329 was testing and why. The tests passed first time. 00:06:12.600 --> 00:06:15.100 So for quickly getting tests set up, it feels 00:06:15.100 --> 00:06:17.240 like handing the task off to a reliable teammate. 00:06:17.480 --> 00:06:20.519 Absolutely. But then test three, the new project 00:06:20.519 --> 00:06:23.120 creation. Ah, this was a deeply disappointing 00:06:23.120 --> 00:06:26.040 failure. Yeah, it really was. The challenge was 00:06:26.040 --> 00:06:29.879 to create a beautiful, modern landing page for 00:06:29.879 --> 00:06:33.779 a fictional brand, AI Fire News. The AI was even 00:06:33.779 --> 00:06:36.449 asked to research the brand's style. the result 00:06:36.449 --> 00:06:41.250 functionally okay maybe but visually basic uninspired 00:06:41.250 --> 00:06:44.050 like something from you know the early 2010s 00:06:44.050 --> 00:06:46.209 and tweaking the prompt adding beautiful modern 00:06:46.209 --> 00:06:49.189 didn't help much so why did it fail so badly 00:06:49.189 --> 00:06:52.209 here well it had no context about design preferences 00:06:52.209 --> 00:06:55.149 no existing code base or design system to learn 00:06:55.149 --> 00:06:58.500 from and the prompt itself beautiful and modern 00:06:58.500 --> 00:07:01.060 is just really subjective and generic for an 00:07:01.060 --> 00:07:03.639 AI. Right. It needs more concrete examples or 00:07:03.639 --> 00:07:05.959 constraints, especially for visual stuff. Exactly. 00:07:06.180 --> 00:07:08.860 So the verdict seems clear. Gemini CLI excels 00:07:08.860 --> 00:07:11.160 at bug fixes, adding features to existing projects, 00:07:11.439 --> 00:07:14.000 refactoring code, analyzing documentation, automated 00:07:14.000 --> 00:07:16.259 testing. But it struggles immensely with new 00:07:16.259 --> 00:07:18.180 projects from scratch that need strong visual 00:07:18.180 --> 00:07:21.120 design, creativity, or aesthetic taste, especially 00:07:21.120 --> 00:07:24.240 without concrete examples or context. Okay. So 00:07:24.240 --> 00:07:26.060 what's the main takeaway from these real -world 00:07:26.060 --> 00:07:28.360 tests then? It shines on existing code bases, 00:07:28.560 --> 00:07:30.800 but struggles with creative design from scratch. 00:07:31.120 --> 00:07:34.120 So getting Gemini CLI running is actually surprisingly 00:07:34.120 --> 00:07:37.459 simple. If you've got Node .js installed, which 00:07:37.459 --> 00:07:39.579 many developers do. Which is that JavaScript 00:07:39.579 --> 00:07:41.899 runtime environment, yeah. Right. You just run 00:07:41.899 --> 00:07:43.939 a single command in your terminal, something 00:07:43.939 --> 00:07:48.000 like npxhttps .github .com, or it's a Google 00:07:48.000 --> 00:07:53.639 Gemini. Conf, the NPS just runs the command directly. 00:07:53.899 --> 00:07:56.319 Then you just type Gemini. First time, it pops 00:07:56.319 --> 00:07:58.180 up in your browser, asks you to log in with your 00:07:58.180 --> 00:08:00.079 personal Google account. And that personal account 00:08:00.079 --> 00:08:02.620 bit is key for the free tier, right? Absolutely 00:08:02.620 --> 00:08:04.839 crucial, yeah. That's what unlocks the generous 00:08:04.839 --> 00:08:08.000 limits. Okay, let's talk power features. Memory 00:08:08.000 --> 00:08:11.120 management. You can actually teach the AI project 00:08:11.120 --> 00:08:14.180 -specific rules. Like memory add, in this project, 00:08:14.259 --> 00:08:16.759 always use npm run lint to check code style. 00:08:17.209 --> 00:08:19.449 Exactly. It helps combat that prompt drift you 00:08:19.449 --> 00:08:21.470 mentioned. Yeah, I still wrestle with prompt 00:08:21.470 --> 00:08:24.649 drift myself sometimes where the AI kind of forgets 00:08:24.649 --> 00:08:26.910 earlier instructions. Giving it concrete rules 00:08:26.910 --> 00:08:28.970 like this seems like a lifesaver. For sure. And 00:08:28.970 --> 00:08:31.850 then there's custom prompts via Gemini .md. This 00:08:31.850 --> 00:08:34.230 is maybe the ultimate context provider. So you 00:08:34.230 --> 00:08:37.269 create a file named Gemini .md in your project 00:08:37.269 --> 00:08:40.110 root. Yep. And in there, using Markdown, you 00:08:40.110 --> 00:08:42.269 write your project context, rules, tech stack, 00:08:42.450 --> 00:08:44.470 preferences, whatever. And Gemini automatically 00:08:44.470 --> 00:08:47.570 loads this context for all interactions, like 00:08:47.570 --> 00:08:49.470 a persistent brain for the project. That's a 00:08:49.470 --> 00:08:52.029 great way to put it. The project's Bible, or 00:08:52.029 --> 00:08:54.370 persistent brain. You might put in hashtag project 00:08:54.370 --> 00:08:57.830 context. This is a React plus Next .js application. 00:08:58.850 --> 00:09:01.690 Hashtag important rules, always run linting, 00:09:01.710 --> 00:09:04.129 stuff like that. That sounds incredibly powerful 00:09:04.129 --> 00:09:06.370 for keeping it on track for complex projects. 00:09:06.789 --> 00:09:08.669 It is, and you can use it interactively in a 00:09:08.669 --> 00:09:10.889 chat session. Just type Gemini or send single 00:09:10.889 --> 00:09:12.950 shot commands directly, like Gemini fix the bug 00:09:12.950 --> 00:09:16.539 in. Flexible. So how does that Gemini .md file 00:09:16.539 --> 00:09:19.340 fundamentally change how the AI operates on a 00:09:19.340 --> 00:09:22.179 project? It provides constant project -specific 00:09:22.179 --> 00:09:25.480 context, dramatically improving AI accuracy and 00:09:25.480 --> 00:09:28.019 relevance. Mid -roll sponsor read provided separately, 00:09:28.179 --> 00:09:30.279 not from source content. All right. So Gemini 00:09:30.279 --> 00:09:32.519 CLI sounds powerful, especially with that Gemini 00:09:32.519 --> 00:09:35.360 .md file, but it doesn't exist in a vacuum, right? 00:09:35.480 --> 00:09:37.399 How does it stack up against the competition? 00:09:37.740 --> 00:09:40.659 Good question. Let's compare. Versus Cursor, 00:09:40.659 --> 00:09:43.059 for instance. Cursor probably wins for that. 00:09:43.340 --> 00:09:47.200 Really deeply integrated IDE experience, especially 00:09:47.200 --> 00:09:48.940 if you're starting a brand new project. Right, 00:09:49.019 --> 00:09:51.100 like a full coding environment built around AI. 00:09:51.399 --> 00:09:54.679 Exactly. But Gemini CLI takes the lead on the 00:09:54.679 --> 00:09:57.200 free usage, that terminal native workflow, which 00:09:57.200 --> 00:09:59.580 some developers prefer, and the unique Google 00:09:59.580 --> 00:10:02.539 search integration. Okay, what about GitHub Copilot? 00:10:02.639 --> 00:10:04.159 That's probably the most well -known competitor. 00:10:04.519 --> 00:10:07.679 Copilot definitely excels at those inline code 00:10:07.679 --> 00:10:10.259 suggestions, the real -time autocomplete as you 00:10:10.259 --> 00:10:13.990 type. It's fantastic for that. But Gemini CLI, 00:10:14.029 --> 00:10:16.269 because it's conversational and has that deep 00:10:16.269 --> 00:10:19.090 context understanding, is often better for higher 00:10:19.090 --> 00:10:21.809 level tasks. Things like refactor this whole 00:10:21.809 --> 00:10:24.149 module or analyze everything in this folder. 00:10:24.309 --> 00:10:26.009 So different strengths for different kinds of 00:10:26.009 --> 00:10:29.049 tasks. And Claude Code? Claude's models sometimes 00:10:29.049 --> 00:10:33.070 seem to have an edge for extremely complex. maybe 00:10:33.070 --> 00:10:35.269 more novel problems that require really sophisticated 00:10:35.269 --> 00:10:38.129 reasoning. But Gemini is generally faster, has 00:10:38.129 --> 00:10:40.669 that huge free tier we keep mentioning, and it 00:10:40.669 --> 00:10:43.350 integrates really smoothly into the wider Google 00:10:43.350 --> 00:10:46.009 ecosystem. So it's a trade -off. Once you get 00:10:46.009 --> 00:10:48.590 the hang of the basics, the source suggests adopting 00:10:48.590 --> 00:10:51.850 more advanced techniques, like an AI -first workflow. 00:10:52.289 --> 00:10:54.049 Yeah, like creating task templates for things 00:10:54.049 --> 00:10:56.990 you do often. And for UI features, providing 00:10:56.990 --> 00:10:59.850 a screenshot of what you want. Apparently a picture 00:10:59.850 --> 00:11:02.049 really is worth a thousand words for the AI here. 00:11:02.210 --> 00:11:04.750 That makes sense. Visual input for visual output. 00:11:05.029 --> 00:11:08.029 Right. And breaking complex features down into 00:11:08.029 --> 00:11:10.509 phases. Right, like building a new feature. First 00:11:10.509 --> 00:11:12.750 do the database schema, then the backend API, 00:11:13.090 --> 00:11:16.169 then the frontend UI, guiding the AI step by 00:11:16.169 --> 00:11:18.429 step. And central to all this seems to be the 00:11:18.429 --> 00:11:21.250 Gemini .md strategy, making it the project's 00:11:21.250 --> 00:11:23.730 single source of truth. Absolutely. The ultimate 00:11:23.730 --> 00:11:26.580 project brain. detailing rules, architecture 00:11:26.580 --> 00:11:29.460 notes, recent changes, keeping the AI fully informed. 00:11:29.679 --> 00:11:32.179 But even with a great co -pilot, things can go 00:11:32.179 --> 00:11:34.960 wrong. Troubleshooting tips. Yeah, if you hit 00:11:34.960 --> 00:11:37.419 rate limiting, just wait a minute or consider 00:11:37.419 --> 00:11:40.299 upgrading. If the code quality seems poor, it's 00:11:40.299 --> 00:11:43.360 almost always a lack of context. So improve your 00:11:43.360 --> 00:11:46.220 Gemini .md or be more specific in the prompt. 00:11:46.559 --> 00:11:49.840 Exactly. Missing dependencies. Point it at your 00:11:49.840 --> 00:11:53.340 package .json. Styling issues. Give it concrete 00:11:53.340 --> 00:11:55.960 examples of components that look right. It's 00:11:55.960 --> 00:11:57.919 all about giving it the right info. This whole 00:11:57.919 --> 00:12:00.759 field moves incredibly fast. What's next for 00:12:00.759 --> 00:12:03.200 tools like this? Expect better design capabilities, 00:12:03.559 --> 00:12:06.759 hopefully addressing that weak spot. More MCP 00:12:06.759 --> 00:12:08.779 server integrations connecting to more tools. 00:12:09.299 --> 00:12:11.500 And probably real -time collaboration features. 00:12:12.080 --> 00:12:14.320 So for developers listening, the key skills seem 00:12:14.320 --> 00:12:17.000 to be prompt engineering, building those comprehensive 00:12:17.000 --> 00:12:20.200 context files and developing AI first workflows. 00:12:20.519 --> 00:12:22.539 That's it. That's how you stay relevant and really 00:12:22.539 --> 00:12:24.559 leverage these tools. What's the single most 00:12:24.559 --> 00:12:26.980 important skill for developers to acquire to 00:12:26.980 --> 00:12:29.919 adapt to this AI powered future? Mastering prompt 00:12:29.919 --> 00:12:32.679 engineering and building robust project context 00:12:32.679 --> 00:12:36.240 files. And this shift. It has really profound 00:12:36.240 --> 00:12:38.360 implications, doesn't it? Yeah, think about solo 00:12:38.360 --> 00:12:40.580 developers. This could mean a huge increase in 00:12:40.580 --> 00:12:43.919 speed, like 10x. Less mental juggling, more time 00:12:43.919 --> 00:12:46.240 for the big picture, the architecture. And for 00:12:46.240 --> 00:12:49.299 teams, faster prototyping, more consistent code 00:12:49.299 --> 00:12:51.620 quality because the AI follows the rules in Gemini 00:12:51.620 --> 00:12:54.700 .md, and dramatically reduced onboarding time 00:12:54.700 --> 00:12:57.159 for new folks. They can just ask the AI about 00:12:57.159 --> 00:13:00.720 the project. Oh, yeah. And for agencies or freelancers, 00:13:00.799 --> 00:13:03.340 deliver projects faster, maybe take on more work. 00:13:03.769 --> 00:13:05.850 Potentially higher profit margins. The source 00:13:05.850 --> 00:13:08.450 material gave Gemini CLI a final overall rating 00:13:08.450 --> 00:13:11.450 of 8 .5 out of 10, which feels about right based 00:13:11.450 --> 00:13:14.009 on the tests. Let's break that down. Simple bug 00:13:14.009 --> 00:13:18.190 fixes. 9 .10. Solid. Complex feature development. 00:13:18.330 --> 00:13:21.490 A perfect 10 .10. Really shines there. New project 00:13:21.490 --> 00:13:24.210 creations, specifically the design part. Only 00:13:24.210 --> 00:13:26.909 5 .10. That's the big weakness. But ease of setup. 00:13:27.529 --> 00:13:30.730 10 .10. And value for money, given the free tier. 00:13:30.950 --> 00:13:33.919 10 .10. So it's a killer tool for almost everything 00:13:33.919 --> 00:13:36.299 except creative design from scratch. That's the 00:13:36.299 --> 00:13:38.899 takeaway. For working on existing code bases, 00:13:39.100 --> 00:13:41.340 understanding deep context, implementing complex 00:13:41.340 --> 00:13:43.740 features, it's revolutionary. The learning curve 00:13:43.740 --> 00:13:46.179 seems minimal, the potential is massive, and 00:13:46.179 --> 00:13:49.519 the price, well, zero for many users is definitely 00:13:49.519 --> 00:13:51.399 right. Yeah, there's really no excuse not to 00:13:51.399 --> 00:13:53.320 give it a try, is there? See how it fits your 00:13:53.320 --> 00:13:56.159 workflow. Tools like Gemini CLI aren't just making 00:13:56.159 --> 00:13:58.440 us incrementally more productive. They feel like 00:13:58.440 --> 00:14:00.399 they're fundamentally changing what it means 00:14:00.399 --> 00:14:03.019 to be a software developer, shifting the focus. 00:14:03.519 --> 00:14:06.580 The choice for you, the learner, seems pretty 00:14:06.580 --> 00:14:09.720 simple. Embrace these tools, learn how to use 00:14:09.720 --> 00:14:11.960 them effectively, multiply your productivity, 00:14:12.200 --> 00:14:15.480 or risk falling behind in this new age of AI. 00:14:15.759 --> 00:14:18.279 So something to mull over. Thinking about your 00:14:18.279 --> 00:14:21.460 next coding project, how might an AI co -pilot 00:14:21.460 --> 00:14:24.559 like Gemini CLI transform your approach, your 00:14:24.559 --> 00:14:26.440 day -to -day? If you're curious to dive deeper 00:14:26.440 --> 00:14:28.740 into this or other areas where AI is making waves, 00:14:28.940 --> 00:14:31.059 you know where to find us. Keep learning, keep 00:14:31.059 --> 00:14:33.080 experimenting. Thanks for deep diving with us 00:14:33.080 --> 00:14:34.379 today. We'll catch you on the next one.